Professor Christensen said that non-rigid image registration (NIR) is an essential tool for making anatomical comparisons. But while many NIR methods have been developed, they are especially difficult to evaluate because there is no known truth or "Gold Standard" that defines how individual points precisely match up from one image to another. His project will attempt to overcome this problem by defining a set of standard tests that can be used to evaluate the performance of NIR methods.
"The knowledge gained through this project will help researchers understand the benefits and limitations of current NIR methods and will provide insights for developing the next generation of methods", Professor Christensen stated. "Our project will develop software tools, provide shared image validation databases for rigorous testing of NRI registration algorithms and extend the scope of prior validation projects."
The goal of the project is to establish, maintain and endorse a standardized set of relevant benchmarks and metrics for performance evaluation of NIR algorithms. The standards will be incorporated into a computer programme to automatically evaluate the registration accuracy of NIR algorithms, according to Professor Christensen, who also serves as the Robert and Virginia Wheeler Faculty Fellow of Engineering and Obermann Scholar.
Nationally known for his research in medical imaging, Professor Christensen has made significant contributions to the fields of medical image registration, deformable shape models and computational anatomy. His medical image registration research will be used to construct anatomical atlases, which can be used to compute average anatomical shape and variability, and will help investigators and medical doctors detect and quantify shape changes due to disease, track the effects of growth, drugs and interventions, and lead the way towards semi-automatic and automatic computer-aided diagnosis, surgery and post-surgical evaluation.
His project co-investigators are Jon Kuhl, Ph.D., professor and departmental executive officer in the UI Department of Electrical and Computer Engineering; Thomas J. Grabowski Jr., M.D., associate professor of neurology and director, Lab of Computational Neuroanatomy in the UI Roy J. and Lucille A. Carver College of Medicine; Hanna Damasio, M.D., University of Southern California professor of psychology and director, Dornseif Cognitive Neuroimaging Center; and Michael W. Vannier, M.D., professor, Department of Radiology, University of Chicago.